Amyotrophic lateral sclerosis (ALS) is the most common form of adult onset motor neuron disease and is characterized by the progressive degeneration and death of motor neurons. The pathologic mechanisms underlying ALS are poorly understood although our laboratory identified decreased levels of transthyretin (TTR), a protein that impacts the retinoid signaling pathway, in the cerebrospinal fluid of ALS patients. Differential expression of retinoid signaling components has been reported in ALS patients and transgenic animal models of familial ALS. We sought to further characterize TTR and retinoid signaling proteins in ALS and to evaluate the role of retinoid signaling in motor neuron cell death. Mass spectrometry and immunoblotting were used to investigate TTR. Immunohistochemistry using lumbar spinal cord tissue from ALS patients and non-neurologic disease controls was used to characterize retinoid signaling pathway proteins. Spinal cord tissue homogenates were used for co-immunoprecipitation studies and electrophoretic mobility shift assays. Motor neuron-enriched cultures established from embryonic day 14 rats were utilized for in vitro studies. RAR-mediated signaling was modulated with pan-agonists and isotype-specific agents and hydrogen peroxide used to model oxidative stress/injury. Altered post-translational modifications and high molecular weight species of the TTR protein were observed in ALS. Cellular retinoic acid binding protein-II (CRABP-II) and retinoic acid receptor beta (RARβ) exhibited increased nuclear localization in motor neurons of sporadic ALS patients. Protein-protein interactions (between CRABP-II and RARα or RARβ) did not differ although retinoic acid response element binding was increased in ALS as compared to controls. Treatment with a pan-RAR or RARβ-specific agonist significantly decreased oxidative stress-induced motor neuron cell death in vitro and genes downstream of RARβ were increased with treatment. Our results indicate that TTR genetic polymorphisms do not represent a novel susceptibility factor for ALS, although protein modification and aggregation appear to be altered in ALS. Localization of proteins of the retinoid signaling pathway is altered in ALS patients and these changes translate to the transcriptional level. Our in vitro work indicates that stimulating the RARs (particularly RARβ) is neuroprotective and that pharmacologic agents that target this nuclear receptor may be of value in slowing the progression of ALS.

Beta-catenin plays multiple roles in liver health and disease through regulation of proliferation, differentiation and metabolism. Elucidating the molecular basis of how beta-catenin regulates these diverse functions and others is the subject of this dissertation. While beta-catenin signaling undergoes temporal activation and its loss dampens liver regeneration (LR), the impact of stimulating this pathway remains unknown. We utilized transgenic (TG) mice expressing Ser45 mutated beta-catenin in hepatocytes to show a growth advantage both in vitro and during LR through cyclin-D1 regulation. Additionally, hydrodynamic delivery of Wnt-1 gene delivery induced beta-catenin activation and hepatocyte proliferation during LR. Regucalcin or senescence marker protein-30 (SMP30) was identified as a beta-catenin target in the liver through the use of hepatocyte-specific beta-catenin conditional knockout (KO) mice. SMP30 is a critical enzyme for the synthesis of ascorbic acid in murine hepatocytes, and its loss led to lower serum ascorbate levels in KO. KO hepatocytes displayed massive apoptosis in culture, which was blocked by addition of ascorbate to culture media. Additionally, apoptosis in HepG2 cells due to regucalcin knockdown was rescued by anti-oxidants. Thus, one mechanism of how beta-catenin regulates hepatocyte redox state and survival is through the control of regucalcin expression. KO livers displayed a basal increase in number of apoptotic hepatocytes. We explored the susceptibility of KO and wildtype (WT) controls to activation of the TNF-alpha mediated apoptotic pathway. Paradoxically, KO mice are refractory to D-galactosamine (GalN)/LPS, Actinomycin D (ActD)/LPS and GalN/TNF-alpha treatments showing lower morbidity than WT. NF-kappaB, a major pro-survival factor and its transcriptional targets were increased in KO basally and after injury due to lack of beta-catenin-p65 association, presence of increased basal inflammation and oxidative stress and increased TLR4 expression in KO livers. Additionally, p65 activation occurred earlier in KO than WT after LPS stimulation. Thus, paradoxical protection from TNF-alpha-mediated apoptosis in KOs occurs owing to pre-existing NF-kappaB activation that 'primes' the liver for protection against exogenous insult. Thus, we have identified beta-catenin as a pleiotropic factor regulating cell proliferation, cellular redox state and cell survival through specific genetic targets and protein-protein interactions. These findings have broad implications in acute and chronic hepatic diseases.

The generation of the pro-inflammatory cytokines IL-6 and TNF-α by macrophages recruited to adipose tissue facilitates obesity-induced inflammation resulting in insulin resistance and type 2 diabetes (T2D). Increased adipose tissue is associated with inflammation and expression of acute phase response (APR) proteins secreted by the liver. Proper homeostasis of the liver is regulated by IL-6-depdendent expression of Hepatocyte Growth Factor (HGF) upon cleavage to its active form (aHGF) by the urokinase-type plasminogen activator (uPA). Plasminogen Activator Inhibitor Type-1 (PAI-1) is a pro-thrombotic APR protein known to inhibit the function of uPA; however, since HGFs activation, interaction and signaling through its receptor, MET are dependent upon uPA, PAI-1 is also capable of regulating the function of hepatic HGF. In vitro data demonstrates that aHGF significantly suppressed IL-6 production by macrophages stimulated with LPS via an increase in phosphorylation of GSK3β, rendering it inactive. Phosphorylated GSK3β correlated with increased retention of the phosphorylated NF-κB p65 subunit in the cytoplasm and an enhanced interaction between CBP and phosphorylated CREB resulting in IL-10 cytokine production. These changes were a direct result of signaling through MET, as effects were reversed in the presence of a selective inhibitor of MET (SU11274) or when using BMM from macrophage-specific conditional MET knockout mice. It is known that obese T2D patients present with an accumulation of PAI-1, which we hypothesize, results in the inactivation of HGF. The loss of HGF-MET signaling results in increased active GSK3β and the progression to unchecked inflammation and disease progression. In vivo studies using male, C57BL6 mice on a high fat diet alongside control fed mice demonstrates move severe hepatic steatosis in obese mice at 44 weeks compared to control. Steatosis coincided with the decrease in aHGF and elevated levels of PAI-1 protein. These results demonstrate that elevated levels of PAI-1 inhibit aHGF, leading to unresolved chronic inflammation in obesity and T2D.

Podocytes are highly differentiated glomerular epithelial cells that play an essential role in the establishment of the glomerular filtration barrier, a structural apparatus that selectively restricts the filtration of different macromolecules in the blood stream on the basis of their sizes, shape and charge. Podocyte dysfunction, one of the major causes of proteinuria, is of pathogenetic and prognostic significance in human glomerular disease. My study is focused on the investigation of novel pathways leading to podocyte dysfunction and proteinuria. In the first part, immunoblotting and quantitative reverse transcriptase PCR (RT-PCR) were used to demonstrate that LIM and senescent cell antigen-like domains 1 (PINCH1) is induced and undergoes nuclear translocation in podocytes after transforming growth factor, beta 1 (TGF-beta1) treatment. Bioinformatics analysis revealed the putative nuclear export signal/nuclear localization signal (NES/NLS) at the PINCH1 C-terminus which is required for its nuclear translocation. Immunoprecipitation and GST pull-down assay identified the interaction between PINCH1 and Wilms tumor 1 (WT1) which led to suppression of the WT1-mediated podocalyxin gene expression. In vivo, PINCH1 also underwent nuclear translocation and interacted with WT1 after TGF-beta1 stimulation. Our data identifies nuclear transcription factor WT1 as a novel binding partner for PINCH1, and provides novel insight into the mechanism of podocyte dysfunction under pathological conditions. In the second part, RT-PCR results revealed that treatment with TGF-beta1 induced gene expression of several wingless-type MMTV integration site family members (Wnts), predominantly Wnt1, and activated beta-catenin in mouse podocytes. Wnt antagonist Dickkopf-1 (DKK1) blocked TGF-beta1-induced beta-catenin activation and preserved nephrin expression. In vivo, ectopic expression of constitutively active TGF-beta1 induced Wnt1 expression, activated glomerular beta-catenin, upregulated its downstream target genes, and led to podocyte injury and proteinuria. Consistently, concomitant expression of DKK1 gene abolished beta-catenin activation in mouse glomeruli, inhibited TGF-beta1-triggered Wnt/beta-catenin target genes, and ameliorated proteinuria. These results establish a role for Wnt/beta-catenin signaling in the pathogenesis of podocyte injury and also suggest that this signaling pathway could be exploited as a therapeutic target for the treatment of proteinuric kidney diseases.

Metabolic networks provide a vital framework for understanding the cellular metabolism in both physiological and pathophysiological states, which will ultimately facilitate network analysis-based drug discovery. In this thesis, we aim to employ a metabolic network analysis approach to study cancer metabolism (a pathophysiological state) and the metabolism of the bacterial pathogen, S. aureus (a physiological state), in order to understand, predict, and ultimately target cell metabolism for drug discovery. Cancer cells have distinct metabolism that highly depend on glycolysis instead of mitochondrial oxidative phosphorylation alone, even in the presence of oxygen, also called aerobic glycolysis or the Warburg effect, which may offer novel therapeutic opportunities. However, the origin of the Warburg effect is only partially understood. To understand the origin of cancer metabolism, our theoretical collaborator, Prof. Alexei Vazquez, developed a reduced flux balance model of human cell metabolism incorporating the macromolecular crowding (MC) constraint and the maximum glucose uptake constraint. The simulations successfully captured the main characteristics of cancer metabolism (aerobic glycolysis), indicating that MC constraint may be a potential origin of the Warburg effect. Notably, when we experimentally tested the model with mammalian cells from low to high growth rates as a proxy of MC alteration, we find that, consistent with the model, faster growing cells indeed have increased aerobic glycolysis. Moreover, the metabolic network analysis approach has also been shown to be capable of predicting the drug targets against pathogen metabolism when completely reconstructed metabolic networks are available. We deduced common antibiotic targets in Escherichia coli and Staphylococcus aureus by identifying shared tissue-specific or uniformly essential metabolic reactions in their metabolic networks. We then predicted through virtual screening dozens of potential inhibitors for several enzymes of these reactions and demonstrated experimentally that a subset of these inhibited both enzyme activities in vitro and bacterial cell viability. Our results indicate that the metabolic network analysis approach is able to facilitate the understanding of cellular metabolism by identifying potential constraints and predicting as well as ultimately targeting the metabolism of the organisms whose complete metabolic networks are available through the seamless integration of virtual screening with experimental validation.

Prostate cancer is the second leading cause of cancer death in men. While prostate specific antigen (PSA) is currently the best biomarker available, its use has many limitations. This study investigates the biosynthesis, secretion and activation of PSA. PSA is secreted as a pro enzyme containing a seven amino acid activation peptide (APLILSR). Because APLILSR is removed extracellularly in vivo, the hypothesis was tested that it may be detected in the blood or urine. Our data indicates that APLILSR is filtered from the bloodstream by the kidney, and is detectable in the urine of patients with prostate cancer, but not controls. Idiopathic pulmonary fibrosis (IPF) is a fatal interstitial lung disease of unknown etiology. Matrix metalloproteinases (MMPs) are a family of proteinases that regulate extracellular matrix turnover and are believed to contribute to IPF. For this reason, the hypothesis that levels of MMP activation peptides will increase in patients with IPF was tested. To test these hypotheses, urine from mice were collected and an ELISA was used to quantify MMP activation peptides. These experiments show that the activation peptides of MMP-2, MMP-7, MMP-8 and MMP-9 are increased in mice with pulmonary fibrosis compared to control mice. The data also showed that that the activation peptides of MMP-2, MMP-7, and MMP-9 are increased in the urine of human patients with IPF compared to healthy controls. These data suggest that urine detection of MMP activation peptides is feasible and correlates with disease. Because urinary detection of the activation peptides of proteinases are indicative of proteinase activation in vivo, the new hypothesis that the accurate measurement of proteinase activation peptides will be relevant clinically arises, and that such measurements may aid in the diagnosis of disease and serve as a marker for following disease progression.

Immunophilins (IP) are receptors for immunosuppressive drugs like cyclosporin A, FK506, rapamycin and their analogs, which are collectively referred to as immunophilin ligands (IPL). The receptors for FK506 belong to the family of FK506- binding proteins (FKBP). Previous studies showed that FK506 augments neuronal growth in vitro. In animal models, the drug promotes morphologic and functional recovery following neuronal lesioning. Here we show for the first time that FK506 has neurotrophic effects in human brain primary cultures. Our data support a possible use of FK506 and its analogues in the treatment of neurodegenerative disorders and as adjuvants in neural tissue transplants. The effect of the drug in vivo, in patients, will ultimately depend on the presence and distribution of IP receptors in the normal and degenerating human brain. FKBP12, the archetypal member of the FKBP family, plays a role in protein folding, protein complexes assembly and intracellular calcium release. Crush injury of facial or sciatic nerves in rat leads to markedly increased FKBP12 levels in the respective nerve nuclei in parallel with nerve regeneration. The presence of FKBP IP has never been demonstrated in the human brain. Using light and immunofluorescent microscopy, laser confocal microscopy and western blotting, we studied FKBP12 expression in a set of archival brain material from Parkinsons disease, Alzheimers disease, dementia with Lewy bodies, encephalitic and non- encephalitic HIV- positive patients and age matched controls. We show that FKBP12 is present in the human brain, predominantly in neurons. Its levels and distribution are altered in the mid- frontal cortex, deep gray matter and midbrain of patients with neurodegenerative diseases. Moreover, it colocalizes with markers of pathology (Lewy bodies, neurofibrillary tangles and neuritic plaques) in areas of neurodegeneration. Disease- specific and region- specific changes are evident. Alterations in basal ganglia FKBP12 levels are also observed in MPTP- treated primates in association with dopaminergic loss (evidenced using PET functional imaging). We propose that the altered expression and distribution of FKBP12 is linked to abnormal protein folding and axonal transport. It may also reflect a compensatory regenerative response that renders immunophilins promising diagnostic and therapeutic targets.

The antioxidant enzyme extracellular superoxide dismutase (EC-SOD) is abundant in the lung and limits inflammation and tissue injury in response to many pulmonary insults. Previous studies reported a loss of full-length EC-SOD from the lung parenchyma with accumulation of proteolyzed EC-SOD in the airspace after interstitial lung injury. However, following airspace only inflammation (pneumonia), EC-SOD accumulates in the airspace without a loss from the interstitium, suggesting this antioxidant may be released from an extrapulmonary source. Because leukocytes are known to express EC-SOD and are prevalent in the bronchoalveolar lavage fluid after injury, it was hypothesized that these cells may transport and release EC-SOD into airspaces. To investigate this, bone marrow chimeras were generated using wild-type and EC-SOD knockout (KO) mice. Following intratracheal treatment with asbestos, reconstituted mice without pulmonary EC-SOD expression, but with EC-SOD in infiltrating and resident leukocytes did not have detectable levels of EC-SOD in the airspaces. In addition, leukocytederived EC-SOD did not significantly lessen inflammation or early stage fibrosis. Although these results indicate that leukocyte-derived EC-SOD is not influential in asbestos-induced interstitial lung injury, EC-SOD in these cells may play a role in attenuating pneumonias and other inflammatory diseases. To test this hypothesis, wild-type and EC-SOD KO mice were given Escherichia coli pneumonia. Notably, even though EC-SOD KO mice had greater pulmonary inflammation than wild-type mice, there was less bacterial clearance from their lungs following infection. While EC-SOD expression has been previously reported in macrophages and neutrophils, its function and subcellular localization in these inflammatory cells is unclear. In this study, EC-SOD was found to be in membrane bound vesicles of phagocytes. This finding led to the hypothesis that inflammatory cell EC-SOD may play a role in antibacterial defense. To investigate this, phagocytes from wild-type and EC-SOD KO mice were evaluated. While macrophages lacking EC-SOD produced more oxidants than EC-SOD expressing cells after stimulation, they had significantly impaired phagocytosis and bacterial killing ability. Overall, these studies suggest that while EC-SOD inside leukocytes does not contribute to interstitial lung injuries, it plays a central role in mediating bacterial infections by facilitating bacterial clearance and limiting inflammation by promoting phagocytosis.

Metastasis is a major contributor to breast cancer mortality, as currently available therapies are unable to ensure progression or disease-free survival. Little is known about the molecular pathogenesis of metastasis, and the role of the surrounding microenvironment is only beginning to be understood. In vitro studies have repeatedly shown that epithelial to mesenchymal transition (EMT) and loss of E-cadherin expression are critical events in the initiation of metastasis and can be induced by the microenvironment. However, metastases are often well-differentiated and epithelial in phenotype, suggesting that EMT is reversible. The role of E-cadherin expression and mesenchymal to epithelial reverting transition (MErT) in metastatic colonization of the secondary site remains ill-defined. Evidence for E-cadherin re-expression and partial MErT was observed in metastases of breast and prostate cancer patients, and suggests that MErT is unstable and reversible. MDA-MB-231 breast cancer cells cultured with hepatocytes also resulted in E-cadherin re-expression and partial MErT, suggesting that such phenotypic plasticity can be induced by the microenvironment of the liver, a key site of breast cancer metastases. Re-expression of E-cadherin following hepatocyte coculture not only results in heterotypic ligation between cancer and liver parenchymal cells, but also activates Erk survival signaling and increases resistance to nutrient-deprivation and chemotherapy. Taken together, our results indicate that the distant organ microenvironment may induce E-cadherin re-expression and partial MErT to enhance the survival of metastatic cancer cells at the secondary organ.

BACKGROUND: Human gliomas account for the most common and malignant tumors in the central nervous system (CNS). Despite optimal treatments, survival of patients with high-grade glioblastoma multiforme (GBM) remains poor. Recent coordinated genomic analyses of a large cohort of clinical GBM specimens identified frequent co-alterations of genes in three core pathwaysthe P53, retinoblastoma (RB), and receptor tyrosine kinase (RTK) pathways that are crucial in gliomagenesis. Further multi-institutional efforts have sub-classified GBMs into four clinical relevant subtypes based on their signature genetic lesions. Among them, PDGFRA overexpression is concomitant with a loss of CDKN2A locus (encoding P16INK4A and P14ARF) in a large number of tumors within one subtype of GBMs. To better understand and design therapeutic strategies against gliomas driven by abnormal platelet-derived growth factor (PDGF) signaling, functional studies using human or mouse models are needed. MAJOR FINDINGS: In order to establish a model that allows us to assess contributions of different signaling pathways to PDGFRα-induced glioma formation, we generated Ink4a/Arf-deficient primary mouse astrocytes (referred to as mAst hereafter) and human glioma cells that overexpress PDGFRα and/or PDGF-A. We found that activation of PDGFRα confers tumorigenicity to Ink4a/Arf-deficient mAst and human glioma cells in the brain. Restoration of p16INK4a but not p19ARF by retroviral transduction suppresses PDGFRα-promoted glioma formation. Mechanistically, abrogation of signaling modules in PDGFRα that lost capacity to bind to SH-2-containing phosphotyrosine phosphatase SHP-2 or Phosphoinositol 3'-Kinase (PI3K) significantly diminished PDGFRα-promoted tumorigenesis. Furthermore, inhibition of SHP-2 by shRNAs or pharmacological inhibitors disrupted the interaction of PI3K with PDGFRα, suppressed downstream AKT/mTOR activation, and impaired tumorigenesis of Ink4a/Arf-null cells, whereas expression of an activated PI3K mutant rescued the effect of SHP-2 inhibition on tumorigenicity. In clinical glioblastoma specimens, PDGFRα and PDGF-A are co-expressed and such co-expression is linked with activation of SHP-2/AKT/mTOR-signaling. Our data thus suggest that in glioblastomas with Ink4a/Arf deficiency, overexpressed PDGFRα promotes tumorigenesis through the PI3K/AKT/mTOR-mediated pathway regulated by SHP-2 activity. SIGNIFICANCE: We expect these findings will improve our understanding of the formation of the gliomas with PDGFRA and INK4A/ARF aberrations. There were studies that predicted SHP-2/PTPN11 as one of the linker genes in clinical GBMs that interact with multiple commonly altered genes. Our results functionally validate this hypothesis and identify SHP-2 as a converge point of several signaling pathways such as PDGFR, EGFR, PI3K, and mTOR that are frequently deregulated in GBMs. It thus represents a promising target for treatments against this fatal disease.